StableReleaseUpdates: gnumed-client (0.2.6.3-1ubuntu0.1) available for testing

[Jordan Mantha]

On 10/22/07, Sarah Hobbs <hobbsee at kubuntu.org> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Michael, what in hell were you thinking?
>
> This is utterly and totally unacceptable.  Gutsy is a stable release,
> and so needs a very high level of QA.  This is precisely *why* we have
> the pain of stable release updates.  You throwing untested crap in there
>  highlights the lack of care you have towards our users.

I'm honestly not sure why you find this such a problem. Often times
testing can be quite complicated and require in-depth knowledge of the
software. If I got a patch from upstream who verified that it worked,
and the package builds and installs fine, I don't see why it shouldn't
be uploaded to -proposed. The whole idea behind having a -proposed
repo is for people to test the SRU, just as Michael has asked.
Personally, I do as much testing as I can but if I'm unable to
reasonable test I ask for help from the community. Is that
unreasonable?

> Part of being a Master of the Universe, (or a Core Dev), is to know what
> is reasonable to go into the archives, and what is not.  An untested
> package is *never* suitable to go into a stable release (or even
> proposed updates to a stable release)!

Untested in what way? I *always* test that an upload builds OK and
installs and I expect the same from other developers, are you wanting
more than that? If you are then I'm sort of left wondering why we even
have -proposed in the first place. It was created so that we had a
place to do the kinds of QA you're taking about.

> At this point, I'd like to question what we do with people who make
> uploads like this.  If these uploads are repeated, I think we really
> need to look at removing their upload rights, because they clearly are
> not suitable for MOTU, based on their lack of care for QA.

This is, IMO, uncalled for and a bit over-reactionary. Michael has a
long history of doing great work for MOTU. Even if we decide that his
actions in this case are not inline with policy, I see, again IMO, no
gross and repeated negligence on his part that would make me question
his MOTU status.

> On the other hand, if this is acceptable conduct for people in MOTU now,
> perhaps we need to document that universe really is completely
> unsupported, and that people should not expect it to mostly work at all.
> If this is the case, I suspect a number of us will seriously consider
> stepping down, or focusing exclusively on main, as MOTU's aims have
> changed, and we actually care about QA.

Honestly, the fact that Michael has even bothered to do an SRU tells
me he's thinking of Universe QA. I'd very much rather we deal with the
issue, clear up any misunderstandings concerning the SRU policy, and
press on.

> I'd like to point out that Michael is not the only one who's been
> pushing recent fixes to gutsy without testing.  These others should also
> read this mail carefully, and think a little more before they upload.

I think indeed we should look at general MOTU education to make sure
all the MOTUs are on the same page and that we have a clear
understanding of policy. I'm convinced, from what I know of Michael,
that he did not knowingly violate any policy, so let's be constructive
here and learn from any mistakes.

> As a candidate for the MOTU council, and the MOTU, I (and the rest of
> MOTU, i suspect) expect better of you.  Please don't do this again.
>
> Hobbsee
>
> Steve Kowalik wrote:
> > Scott Kitterman wrote:
> >> I'm sending this IRC snippet (my question on #ubuntu-motu) was after reading
> >> the above mail?
> >>
> >> [15:49] <ScottK> geser: Please tell me you didn't upload a fix to
> >> gutsy-proposed that you haven't verified works (that's what I get from your
> >> mail to the MOTU list)?
> >> [15:50] <geser> ScottK: I didn't check the package myself
> >> [15:50] <ScottK> geser: Did you upload it?
> >> [15:50] <geser> yes
> >> [15:50] <geser> ScottK: that patch comes from upstream
> >> [15:51] * ScottK sort of thought testing before uploading would have been a
> >> good idea?
> >>
> >> I'd like a clarification of policy here (I've thought I knew for sure the
> >> right answer and been wrong before).  I thought SRUs were supposed to be
> >> tested before uploading to *-proposed.  Is that wrong?
> >
> > Certainly not. The thing to keep in mind is that you are updating a
> > *STABLE* release -- if you throw untested broken crap there, people
> > aren't going to be very happy -- at the very least make sure what you're
> > uploading builds, works, and doesn't have any regressions.
> >
> > Cheers,
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
>
> iD8DBQFHHYS37/o1b30rzoURAs6HAKCuJFnIXi7BR5ZR3aNDAjP8LE1fVACeMxhp
> fWhCVqISyxtJAmEjXqQqOyw=
> =+x16
> -----END PGP SIGNATURE-----
>
> --
> Ubuntu-motu mailing list
> Ubuntu-motu at lists.ubuntu.com
> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu
>
>